Our world is filled with huge amount of data streams that are generated by hardware/software sensors. Such streams, containing information on air pollution, road conditions, amount of traffic, human movement, and social web data are source of what we term "Big Data." This project is developing an infrastructure which heterogeneous data streams are uniformly managed, selectively aggregated, and visualized for people to utilize them. We present the infrastructure and the applications, called SmartCityApps and Sensorizer, and show our contribution to solve personal, local governmental and national problems in the real world.

Cities have been experiencing emerging challenges such as efficient energy management, economic growth and development, security and quality of life of its habitants. ClouT's overall concept is leveraging the Cloud Computing as an enabler to bridge the Internet of Things with Internet of People via Internet of Services, to establish an efficient communication and collaboration platform exploiting all possible information sources to make the cities smarter. This project is a part of EU-Japan projects and it collaborates with EU and JP consortium members including 4 cities (Santander, Genova, Fujisawa and Mitaka).

We envision Social Big Data World where huge amount of data generated by a variety of sensors are consumed by a huge number of services simultaneously. This project aims at establishing an infrastructural technology that makes the data circulate among consumers, enabling the consumers to integrate, amplify, learn and predict the world from the data. A number of potential applications on it will provide the power of real-time city management with functionalities of, for example, real world event detection/classification, N-dimensional analysis and visualization of cities, and city resources control.

A cloud system in general is a complex system that is constructed with a number of server computers, network appliances, and disk storage devices. A variety of stakeholders depend on the system to provide services to their customers. This research aims at assuring dependability of cloud systems by defining dependability requirements in detail, deciding evidences needed to assure the requirements satisfaction, and executing them in D-Case, a dependability assurance tool.

According to increase number of citizens in city area, creating various services such as navigation, recommendation and effectiveness are required by detecting various social events in city area. In this research, we present a system which detects and classifies social events by analyzing various data which are posted to social network and social graph. In addition, we evaluate effectiveness of our system by presenting a user interface which presents detected various events.

This project aims to establish a user-centric services infrastructure, in which bits from users mind are circulated locally via user-side terminals and place services to approach to the users to help their activities. We demonstrate an application that leverages Bluetooth to circulate the mind bits between user-side terminals and public displays, and how such an application can help user activities.

The aim of our project is to predict year 2020's new emerged needs, and indicating services to fulfill the needs. In addition, we create a prototype of application which realizes the services, and conduct experiments with actual cars. In this exhibition, we introduce AwareCar technology, which realizes more human-like automatic driving system by using inner-car and outer-car information based on various sensors.

We develop a distributed middleware system that can accommodate global-scale communication among networked things and applications. The things include sensors and actuators that interface between the physical world and the cyber world, and processors that manipulate the data flows among sensors and actuators. Functional requirements of the system are the following two-fold. First, the system needs to enable applications to discover and lookup things in IoE under scopes with different granularity, i.e., global, cloud, and fog, and also based on the things' functionality. Second, The system needs to enable applications to communicate with things, and make them communicate one another. We use SOX (sensor over XMPP) protocol for exchanging hard/soft sensor/actuator data as well as meta data.

Communication among users is activated by "the common topics." However, it is getting difficult to have the common topics among us since information is provided based on each user's individual interest in the recent situation with diversification of the media, as the Internet has been ubiquitous. To solve this problem, T-ras extracts user's interests and recommends the TV programs that are popular on Twitter before its actual broadcasting. Also T-ras recommends a user the programs that other users with similar preferences will watch and supplies common topics. User's communication will be activated by using T-ras.

In recent years, soft-bodied robots composed of soft materials have been explored against conventional rigid-bodied robots. Soft robots aim to achieve wide adaptivity by copying natural movements of living organisms, and had been very successful so far. However, methods of operating those hardwares have not progressed as rapid as its hardware design. In this research, we show what kind of behavior is suitable for soft robots, by using 3D-printed, polygon-shaped hardware as one example of soft robots.

By the development of the Information technology, data accumulated from sensors are now utilized to make effects in the real space. However, the behavior and the needs of people changes drastically, and the detection of it is still making slow progress. Our research aims to detect the context of the real space, and to present suitable interaction to reach the analyzed data.

There are telepresence systems that allow a person to provide sense of sharing the same space remotely, such as TV conference. In these systems, sense of sharing the same space is important. However, current systems do not satisfy such sense because of conversation via screen or planer sound. In my research, I focus on the sound that takes an important role to feel like sharing the same space remotely. Knowing that the person who you are talking with or the place where sounds are coming from, we can feel a sign of a person's existing unconsciously. There are some factors of sound to feel the sign. Among them, I propose a new telepresence sound system which provides the "sound of location" and "direction of conversation."

Recently, contexts which are not raw number data like temperature and humidity are aware as sensor data. Whether you have an empty space on the parking space is one good example. On the other hand, on sensor networks, lots of work has contributed in this field of interpolating lost data. Although, main assumptions on existing researches are only for environmental monitoring data, which values take similar consequence in similar time and similar place. On this research, we interpolate lost sensor value which could not resolve on existing research's assumptions.

Recently, people using web to search for cooking recipe is increasing. In one of the most famous Japanese recipe web site "CookPad," about 2 millions of recipes have been contributed, and they have rich contents of recipe including its name, foodstuffs, purpose of that recipe, and etc. Although, by the combination of foodstuffs, user cannot find out what they really want. On this research, we propose and implement an application, which creates recipes automatically, from foodstuffs and keywords. We use machine learning by extracting feature quantity from recipe data which CookPad is exposing.